|
{ |
|
"best_metric": 23.816679188580014, |
|
"best_model_checkpoint": "../models/whisper-small-kn/checkpoint-5000", |
|
"epoch": 0.3, |
|
"global_step": 5000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 4.960000000000001e-06, |
|
"loss": 1.2587, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 9.960000000000001e-06, |
|
"loss": 0.4126, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_loss": 2.279691219329834, |
|
"eval_runtime": 162.3379, |
|
"eval_samples_per_second": 0.616, |
|
"eval_steps_per_second": 0.025, |
|
"eval_wer": 127.26387536514119, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 9.44888888888889e-06, |
|
"loss": 0.263, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 8.893333333333333e-06, |
|
"loss": 0.2099, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_loss": 0.1774223893880844, |
|
"eval_runtime": 162.1564, |
|
"eval_samples_per_second": 0.617, |
|
"eval_steps_per_second": 0.025, |
|
"eval_wer": 28.249436513899322, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 8.337777777777777e-06, |
|
"loss": 0.1891, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 7.782222222222223e-06, |
|
"loss": 0.1736, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"eval_loss": 0.15647487342357635, |
|
"eval_runtime": 163.7282, |
|
"eval_samples_per_second": 0.611, |
|
"eval_steps_per_second": 0.024, |
|
"eval_wer": 27.573253193087904, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 7.226666666666667e-06, |
|
"loss": 0.1616, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 6.671111111111112e-06, |
|
"loss": 0.1506, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"eval_loss": 0.15138228237628937, |
|
"eval_runtime": 161.6686, |
|
"eval_samples_per_second": 0.619, |
|
"eval_steps_per_second": 0.025, |
|
"eval_wer": 26.033057851239672, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 6.1155555555555555e-06, |
|
"loss": 0.1452, |
|
"step": 2250 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 5.560000000000001e-06, |
|
"loss": 0.1373, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"eval_loss": 0.1493964046239853, |
|
"eval_runtime": 164.5271, |
|
"eval_samples_per_second": 0.608, |
|
"eval_steps_per_second": 0.024, |
|
"eval_wer": 24.417731029301276, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 5.004444444444445e-06, |
|
"loss": 0.133, |
|
"step": 2750 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 4.448888888888889e-06, |
|
"loss": 0.1298, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"eval_loss": 0.14558260142803192, |
|
"eval_runtime": 165.4317, |
|
"eval_samples_per_second": 0.604, |
|
"eval_steps_per_second": 0.024, |
|
"eval_wer": 25.05634861006762, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 3.893333333333333e-06, |
|
"loss": 0.1249, |
|
"step": 3250 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"learning_rate": 3.337777777777778e-06, |
|
"loss": 0.1198, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"eval_loss": 0.14362207055091858, |
|
"eval_runtime": 165.6164, |
|
"eval_samples_per_second": 0.604, |
|
"eval_steps_per_second": 0.024, |
|
"eval_wer": 24.417731029301276, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 2.7822222222222222e-06, |
|
"loss": 0.1103, |
|
"step": 3750 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 2.226666666666667e-06, |
|
"loss": 0.1102, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_loss": 0.14516383409500122, |
|
"eval_runtime": 163.9446, |
|
"eval_samples_per_second": 0.61, |
|
"eval_steps_per_second": 0.024, |
|
"eval_wer": 24.267468069120962, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 1.6711111111111112e-06, |
|
"loss": 0.1113, |
|
"step": 4250 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 1.1155555555555558e-06, |
|
"loss": 0.1097, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"eval_loss": 0.14023457467556, |
|
"eval_runtime": 164.1886, |
|
"eval_samples_per_second": 0.609, |
|
"eval_steps_per_second": 0.024, |
|
"eval_wer": 24.30503380916604, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 5.6e-07, |
|
"loss": 0.1083, |
|
"step": 4750 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 4.444444444444445e-09, |
|
"loss": 0.105, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"eval_loss": 0.13982167840003967, |
|
"eval_runtime": 162.4084, |
|
"eval_samples_per_second": 0.616, |
|
"eval_steps_per_second": 0.025, |
|
"eval_wer": 23.816679188580014, |
|
"step": 5000 |
|
} |
|
], |
|
"max_steps": 5000, |
|
"num_train_epochs": 9223372036854775807, |
|
"total_flos": 9.234646234619904e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|